CN113204235A - Automatic obstacle avoidance algorithm of fire extinguishing robot - Google Patents
Automatic obstacle avoidance algorithm of fire extinguishing robot Download PDFInfo
- Publication number
- CN113204235A CN113204235A CN202110396197.5A CN202110396197A CN113204235A CN 113204235 A CN113204235 A CN 113204235A CN 202110396197 A CN202110396197 A CN 202110396197A CN 113204235 A CN113204235 A CN 113204235A
- Authority
- CN
- China
- Prior art keywords
- fire
- fuzzy
- obstacle
- distance
- obstacle avoidance
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000013528 artificial neural network Methods 0.000 claims abstract description 53
- 238000012545 processing Methods 0.000 claims abstract description 27
- 238000000034 method Methods 0.000 claims abstract description 24
- 230000004888 barrier function Effects 0.000 claims abstract description 8
- 230000006870 function Effects 0.000 claims description 19
- 238000012549 training Methods 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 6
- 238000013461 design Methods 0.000 claims description 3
- 238000013178 mathematical model Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 3
- 238000013507 mapping Methods 0.000 description 4
- 230000033001 locomotion Effects 0.000 description 3
- 210000002569 neuron Anatomy 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000010187 selection method Methods 0.000 description 2
- 230000001131 transforming effect Effects 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000002238 attenuated effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005316 response function Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000004613 tight binding model Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0223—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving speed control of the vehicle
Abstract
The invention discloses an automatic obstacle avoidance algorithm of a fire-extinguishing robot, which has the technical scheme key points that: the method comprises the following steps: acquiring position information of an obstacle in a field environment relative to the fire-extinguishing robot by arranging sensors to obtain an input obstacle distance D of an obstacle avoidance algorithm; fuzzification processing is carried out on the barrier distance D and the left and right wheel speed values of the fire-extinguishing robot, the input barrier distance and the left and right wheel speed values of the fire-extinguishing robot are converted into fuzzy language values, and the fuzzy language value variables represent fuzzy subsets in a specific theoretical domain; the obstacle avoidance method of the automatic obstacle avoidance algorithm of the fire-extinguishing robot adopts fuzzy description to complete the behavior coding of the distance of the obstacle relative to the fire-extinguishing robot and the speed of the left wheel and the right wheel of the fire-extinguishing robot, simultaneously fully utilizes the self-learning, nonlinear approximation and self-adaption technologies of the RBF neural network, and realizes the obstacle avoidance rule under the unknown environment through the information input of the sensor and the modeling of the internal fuzzy neural network environment.
Description
Technical Field
The invention belongs to the fields of robots, artificial intelligence and the like, and particularly relates to an automatic obstacle avoidance algorithm of a fire-extinguishing robot.
Background
With the development and progress of society, various modern electric devices are present in our lives. Therefore, the occurrence of fire has become increasingly frequent in our lives, and the work of fire fighters has become increasingly difficult. The disaster site is full of various dangers, and extremely serious threat can be caused to the personal safety of fire-fighting officers, so that the research of the fire-fighting robot has very important research value for social safety, the danger index of fire-fighting rescue actions is reduced, casualties are effectively avoided or reduced, the fire-fighting speed is increased, the efficiency is improved, and the social property is guaranteed. And the intelligent obstacle avoidance of the fire-extinguishing robot during the traveling process directly relates to whether the fire-extinguishing robot can safely reach a fire-extinguishing point and complete a task.
Reference may be made to chinese patent publication No. CN111408089A, which discloses a fire-fighting robot and a fire-fighting robot fire-extinguishing system, the fire-fighting robot comprising: the chassis driving system is used for navigating and walking the fire-fighting robot; the navigation sensing system is used for navigating under the control of the control system and automatically positioning the fire fighting robot; the communication system is used for sending the status information and the fire information of the fire-fighting robot to the background system through a wireless network and forwarding a control instruction sent by the background system to the control system; the flame identification system is used for identifying and ranging flames of fire sources in a certain range around the fire sources through an image algorithm, positioning the fire sources and sending fire information through the communication system; the fire extinguishing system is used for extinguishing fire according to a fire extinguishing instruction under the control of the control system, and the charging station system is used for charging the fire-fighting robot under the control of the control system; and the control system is used for controlling the cooperative operation of all systems in the fire-fighting robot. The system does not deeply solve the technical problem of automatic obstacle avoidance from the aspect of algorithm.
Different algorithms such as an artificial market method, a fuzzy control method, a neural network method and the like can be selected when the fire-extinguishing robot carries out obstacle avoidance in motion, but the real-time performance or the accuracy of the algorithms is not high enough. Therefore, the patent provides a novel intelligent obstacle avoidance algorithm of the fire-extinguishing robot, namely a fuzzy RBF neural network algorithm combining a fuzzy rule and a neural network algorithm.
Disclosure of Invention
In view of the problems mentioned in the background art, the invention aims to provide an automatic obstacle avoidance algorithm of a fire-extinguishing robot, so as to solve the problems mentioned in the background art.
The technical purpose of the invention is realized by the following technical scheme:
an automatic obstacle avoidance algorithm for a fire fighting robot, comprising:
acquiring position information of an obstacle in a field environment relative to the fire-extinguishing robot by arranging sensors to obtain an input obstacle distance D of an obstacle avoidance algorithm;
fuzzification processing is carried out on the barrier distance D and the left and right wheel speed values of the fire-extinguishing robot, the input barrier distance and the left and right wheel speed values of the fire-extinguishing robot are converted into fuzzy language values, and the fuzzy language value variables represent fuzzy subsets in a specific theoretical domain and correspond to certain membership functions;
designing a fuzzy neural network control rule, and establishing a control rule between the azimuth distance of the obstacle relative to the fire-fighting robot and a fuzzy set for controlling the left and right wheel speeds of the fire-fighting robot;
constructing a structure of a fuzzy control RBF neural network, training and adjusting weights of the fuzzy control RBF neural network, and constructing a mathematical model between an obstacle distance fuzzy set and a left and right wheel speed fuzzy set of the fire-fighting robot;
and performing defuzzification processing on the fuzzy quantity of the speeds of the left wheel and the right wheel of the fire-extinguishing robot to obtain the speed numerical values of the left wheel and the right wheel of the fire-extinguishing robot.
Preferably, the obstacle avoidance control process of the fire-extinguishing robot includes:
starting;
initializing a system;
collecting distance and speed information;
fuzzification of distance amount and speed;
fuzzy rule processing;
defuzzification processing;
setting the speed of the left wheel and the right wheel;
and (6) displaying.
Preferably, the training and adjusting of the RBF neural network weight are performed between the distance quantity, the speed fuzzification and the fuzzy rule processing steps, and the defuzzification processing is performed after the training and adjusting of the RBF neural network weight.
Preferably, the step of feeding back the value set by the left and right wheel speeds to the acquisition distance and speed information is performed between the step of setting the left and right wheel speeds and the step of displaying.
Preferably, the input distance D includes three input distance parameters, namely, a left front distance D1, a right front distance D2 and a right front distance D3 of the obstacle with respect to the fire-fighting robot, where the input distance D information is { D1, D2 and D3}, and the fuzzy languages thereof are: far, near, and near.
Preferably, the output speed parameters in the left and right wheel speed values of the fire-fighting robot are the rotation speeds of the left wheel and the right wheel of the fire-fighting robot respectively, that is, the information of the speed V is { V1, V2}, the speed is blurred to be 5 levels { NB, NS, Z, PS, PB }, and a certain speed value corresponding to each speed value is selected as a membership function of the speed value.
Preferably, when designing the fuzzy neural network control rule, the fuzzy neural network control rule is first established in an experience knowledge and expert base, and obstacle avoidance conditions are classified into 9 classes according to an obstacle avoidance environment of the fire fighting robot, wherein a non-obstacle is classified into class 1, a front obstacle is classified into class 2, a left obstacle is classified into class 3, a right obstacle is classified into class 4, front and right obstacles are classified into class 5, front and left obstacles are classified into class 6, left, front and left obstacles are classified into class 7, right, front and right obstacles are classified into class 8, and left and right obstacles are classified into class 9, and 189 control rules are provided.
Preferably, the structural design of the fuzzy-control RBF neural network comprises: selecting three layers of RBF neural networks as neural networks of fuzzy control rules, taking input nodes of the RBF neural networks as distance parameters, taking output nodes of the RBF neural networks as speed parameters, and determining fuzzy magnitude values of input and output parameters according to membership functions of fuzzy subsets.
Preferably, when the defuzzification processing is performed, defuzzification processing is performed on the output of the fuzzy neural network, and an area center method is adopted for a defuzzification algorithm.
In summary, the invention mainly has the following beneficial effects:
the obstacle avoidance method of the automatic obstacle avoidance algorithm of the fire-extinguishing robot adopts fuzzy description to complete the behavior coding of the distance of the obstacle relative to the fire-extinguishing robot and the speed of the left wheel and the right wheel of the fire-extinguishing robot, simultaneously fully utilizes the self-learning, nonlinear approximation and self-adaption technologies of the RBF neural network, and realizes the obstacle avoidance rule under the unknown environment through the information input of the sensor and the modeling of the internal fuzzy neural network environment. The obstacle avoidance algorithm of the fuzzy RBF neural network is adopted, so that the processing capacity of the system on the data acquired by the sensor is greatly improved, and an accurate obstacle avoidance decision can be made according to the obstacle avoidance environment.
Drawings
FIG. 1 is a block flow diagram of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1
Referring to fig. 1, an automatic obstacle avoidance algorithm of a fire-fighting robot includes:
acquiring position information of an obstacle in a field environment relative to the fire-extinguishing robot by arranging sensors to obtain an input obstacle distance D of an obstacle avoidance algorithm;
fuzzification processing is carried out on the barrier distance D and the left and right wheel speed values of the fire-extinguishing robot, the input barrier distance and the left and right wheel speed values of the fire-extinguishing robot are converted into fuzzy language values, and the fuzzy language value variables represent fuzzy subsets in a specific theoretical domain and correspond to certain membership functions;
designing a fuzzy neural network control rule, and establishing a control rule between the azimuth distance of the obstacle relative to the fire-fighting robot and a fuzzy set for controlling the left and right wheel speeds of the fire-fighting robot;
constructing a structure of a fuzzy control RBF neural network, training and adjusting weights of the fuzzy control RBF neural network, and constructing a mathematical model between an obstacle distance fuzzy set and a left and right wheel speed fuzzy set of the fire-fighting robot;
and performing defuzzification processing on the fuzzy quantity of the speeds of the left wheel and the right wheel of the fire-extinguishing robot to obtain the speed numerical values of the left wheel and the right wheel of the fire-extinguishing robot.
Referring to fig. 1, the obstacle avoidance method of the automatic obstacle avoidance algorithm of the fire-fighting robot adopts fuzzy description to complete the behavior coding of the distance of the obstacle relative to the fire-fighting robot and the speed of the left wheel and the right wheel of the fire-fighting robot, simultaneously fully utilizes the self-learning, nonlinear approximation and adaptive technologies of the RBF neural network, and realizes the obstacle avoidance rule under the unknown environment through the information input of the sensor and the modeling of the internal fuzzy neural network environment. The obstacle avoidance algorithm of the fuzzy RBF neural network is adopted, so that the processing capacity of the system on the data acquired by the sensor is greatly improved, and an accurate obstacle avoidance decision can be made according to the obstacle avoidance environment.
Referring to fig. 1, the obstacle avoidance control process of the fire-fighting robot includes:
starting;
initializing a system;
collecting distance and speed information;
fuzzification of distance amount and speed;
fuzzy rule processing;
defuzzification processing;
setting the speed of the left wheel and the right wheel;
displaying;
and carrying out RBF neural network weight training and adjusting between the distance quantity, speed fuzzification and fuzzy rule processing steps, and carrying out defuzzification processing after the RBF neural network weight training and adjusting step.
Wherein, the step of feeding back the numerical value set by the left and right wheel speeds to the acquisition distance and speed information is carried out between the step of setting the left and right wheel speeds and the step of displaying.
Referring to FIG. 1, at the beginning of the pass; initializing a system; collecting distance and speed information; fuzzification of distance amount and speed; fuzzy rule processing; defuzzification processing; setting the speed of the left wheel and the right wheel; after the steps of displaying and the like, the obstacle avoidance capability of the fire-fighting robot can be improved well, and the robot is ensured to safely reach a fire-fighting point at a set speed and complete a task.
Referring to fig. 1, wherein upon system initialization, the fire fighting robot is configured with a camera configured to take a first image of an initialization object when the fire fighting robot is in a first position and a second image of the initialization object when the robot is in a second position; an odometer unit configured to determine a position of the fire-fighting robot relative to the initialization object; and a controller configured to: initializing the odometer unit with respect to an initialization object determined at least in part from the first image when the fire fighting robot is in a first position, demonstrating a learning route by the user starting from the first position, wherein the learning route associates a motion of the robot with a position of the fire fighting robot with respect to the initialization object determined by the odometer unit, initializing the odometer unit with respect to the initialization object determined at least in part from the second image when the fire fighting robot is in a second position, and autonomously walking along at least a portion of the learning route from the second position when the fire fighting robot is instructed to perform one or more associated motions based at least in part on the position of the fire fighting robot with respect to the initialization object determined by the odometer unit.
Referring to fig. 1, in acquiring distance and speed information, a distance acquisition module and a speed acquisition module are used, which can accurately acquire the distance to an obstacle and the related speed information.
Referring to fig. 1, there are three input distance parameters in the input amount obstacle distance D, which are the distance D1 to the left front, the distance D2 to the right front, and the distance D3 to the right front of the obstacle, i.e. the information of the input distance D is { D1, D2, D3}, and the fuzzy languages thereof are: far, near, and near.
Referring to fig. 1, the output speed parameters in the left and right wheel speed values of the fire-fighting robot are the rotation speeds of the left and right wheels of the fire-fighting robot, i.e. the information of the speed V is { V1, V2}, the speed is blurred to 5 levels { NB, NS, Z, PS, PB }, and a certain speed value corresponding to each speed value is selected as a membership function of the speed metric.
Referring to fig. 1, when designing the fuzzy neural network control rule, empirical knowledge and an expert base are first established, obstacle avoidance conditions are classified into 9 classes according to an obstacle avoidance environment of the fire fighting robot, wherein an obstacle-free object is set as class 1, a front obstacle is set as class 2, a left obstacle is set as class 3, a right obstacle is set as class 4, front and right obstacles are set as class 5, front and front left obstacles are set as class 6, left, front left and front obstacles are set as class 7, right, front right and front obstacles are set as class 8, and left and right obstacles are set as class 9, and 189 control rules are provided.
Referring to fig. 1, wherein the fuzzy-control RBF neural network structure design includes: selecting three layers of RBF neural networks as neural networks of fuzzy control rules, taking input nodes of the RBF neural networks as distance parameters, taking output nodes of the RBF neural networks as speed parameters, and determining fuzzy magnitude values of input and output parameters according to membership functions of fuzzy subsets.
Referring to fig. 1, in the defuzzification process, the defuzzification process is performed on the output of the fuzzy neural network, and the defuzzification algorithm adopts an area center method.
Wherein, the value belonging to 0, 1 in the fuzzy logic is used for representing the degree, is an extension of the binary logic, and is suitable for describing the inaccuracy in the actual life. The key concepts are: a gradual membership. A collection may have elements that partially belong to it; a proposition may be both, and there is partial genuine. Fuzzy logic is a method and tool for representing and analyzing uncertain, inaccurate information by mimicking human thinking. Fuzzy logic is not fuzzy in itself, it is not "fuzzy" logic, but rather is logic that handles "fuzziness" (phenomena, events) to the effect that the fuzziness is removed. The character of a concept is described by using the value of a membership function, that is, the degree of a concept belonging to an element is represented by a numerical value between 0 and 1, and the value is called the membership degree of the element pair set. When the membership is 1 or 0, it is like the "true" and false "in the conventional mathematics, and when there is a difference between them, it belongs to the gray zone between true and false. Fuzzy sets are sets that are indistinct in boundaries or boundaries and have certain things to establish membership functions to represent fuzzy sets.
Wherein, the RBF is a three-layer forward network with a single hidden layer. The first layer is an input layer and consists of signal source nodes. The second layer is a hidden layer, the number of nodes of the hidden layer depends on the needs of the described problem, the transformation function of the neuron in the hidden layer, namely the radial basis function, is a non-negative linear function which is radially symmetrical and attenuated to the central point, the function is a local response function, and the specific local response is reflected in that the transformation from the visible layer to the hidden layer is different from other networks. Previous forward network transformation functions were all functions of the global response. The third layer is the output layer, which is responsive to the input mode. The input layer only plays a role of transmitting signals, the input layer and the hidden layer can be regarded as connection with the connection weight value of 1, and the tasks completed by the output layer and the hidden layer are different, so that the learning strategies of the output layer and the hidden layer are different. The output layer adjusts the linear weight and adopts a linear optimization strategy, so that the learning speed is high; the hidden layer adjusts parameters of an activation function (green function, gaussian function, the latter is generally taken), and a nonlinear optimization strategy is adopted.
The basic idea of the RBF neural network: and (3) forming a hidden layer space by using RBF as a base of a hidden unit, transforming an input vector by using the hidden layer, and transforming low-dimensional mode input data into a high-dimensional space, so that the problem of linear inseparability in the low-dimensional space is linearly separable in the high-dimensional space. In detail, the hidden layer space is formed by the base of the hidden unit of the RBF, so that the input vector can be directly mapped to the hidden space (not connected by the weight). When the center point of the RBF is determined, the mapping relation is determined. The mapping from the hidden layer space to the output space is linear (note that the relationship between linear mapping and non-linear mapping is distinguished), that is, the network output is the linear weighted sum of the unit outputs, and the weight here is the network adjustable parameter.
The RBF neural network center selection method comprises the following steps:
for the learning algorithm of the RBF neural network, the key problem is the reasonable determination of the neuron center parameters of the hidden layer. The common method is to select the central parameter (or its initial value) from a given training sample set directly according to a certain method, or to determine it by using a clustering method.
Direct calculation method (random selection RBF center)
The centers of the hidden layer neurons are randomly chosen among the input samples and the centers are fixed. Once the center is fixed, the output of the hidden layer neurons is known and the connection weights of such a neural network can be determined by solving a system of linear equations. The distribution applicable to the sample data is clearly representative.
Second, RBF center selection method for self-organizing learning
The center of the RBF neural network can vary and its location is determined by self-organizing learning. The linear weights of the output layers are determined by supervised learning. The method is to redistribute the neural network resources and make the neuron center of the hidden layer of the RBF located in the important area of the input space through learning. The method mainly adopts a K-means clustering method to select the center of the RBF.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (9)
1. The utility model provides an automatic obstacle avoidance algorithm of fire extinguishing robot which characterized in that: the method comprises the following steps:
acquiring position information of an obstacle in a field environment relative to the fire-extinguishing robot by arranging sensors to obtain an input obstacle distance D of an obstacle avoidance algorithm;
fuzzification processing is carried out on the barrier distance D and the left and right wheel speed values of the fire-extinguishing robot, the input barrier distance and the left and right wheel speed values of the fire-extinguishing robot are converted into fuzzy language values, and the fuzzy language value variables represent fuzzy subsets in a specific theoretical domain and correspond to certain membership functions;
designing a fuzzy neural network control rule, and establishing a control rule between the azimuth distance of the obstacle relative to the fire-fighting robot and a fuzzy set for controlling the left and right wheel speeds of the fire-fighting robot;
constructing a structure of a fuzzy control RBF neural network, training and adjusting weights of the fuzzy control RBF neural network, and constructing a mathematical model between an obstacle distance fuzzy set and a left and right wheel speed fuzzy set of the fire-fighting robot;
and performing defuzzification processing on the fuzzy quantity of the speeds of the left wheel and the right wheel of the fire-extinguishing robot to obtain the speed numerical values of the left wheel and the right wheel of the fire-extinguishing robot.
2. The automatic obstacle avoidance algorithm of the fire extinguishing robot according to claim 1, characterized in that: the obstacle avoidance control process of the fire-extinguishing robot comprises the following steps:
starting;
initializing a system;
collecting distance and speed information;
fuzzification of distance amount and speed;
fuzzy rule processing;
defuzzification processing;
setting the speed of the left wheel and the right wheel;
and (6) displaying.
3. The automatic obstacle avoidance algorithm of the fire extinguishing robot according to claim 2, characterized in that: and training and adjusting the RBF neural network weight between the distance quantity, the speed fuzzification and the fuzzy rule processing step, and performing the defuzzification processing after the RBF neural network weight training and adjusting step.
4. The automatic obstacle avoidance algorithm of the fire extinguishing robot according to claim 2, characterized in that: and feeding back the numerical value set by the left and right wheel speeds to the acquisition distance and speed information step between the left and right wheel speed setting step and the display step.
5. The automatic obstacle avoidance algorithm of the fire extinguishing robot according to claim 1, characterized in that: the input distance D includes three input distance parameters, namely, a left front distance D1, a right front distance D2, and a right front distance D3 of the obstacle with respect to the fire-fighting robot, that is, the information of the input distance D is { D1, D2, D3}, and the fuzzy languages thereof are: far, near, and near.
6. The automatic obstacle avoidance algorithm of the fire extinguishing robot according to claim 1, characterized in that: the output speed parameters in the left and right wheel speed values of the fire-fighting robot are respectively the rotating speeds of the left wheel and the right wheel of the fire-fighting robot, namely the information of the speed V is { V1, V2}, the speed is blurred to be 5 grades { NB, NS, Z, PS, PB }, the speeds are respectively corresponding to certain speed values, and a Gaussian function is selected as a membership function of the speed values.
7. The automatic obstacle avoidance algorithm of the fire extinguishing robot according to claim 1, characterized in that: when the fuzzy neural network control rule is designed, experience knowledge and an expert base are firstly established, obstacle avoidance conditions are divided into 9 types according to the obstacle avoidance environment of the fire-fighting robot, wherein the obstacle-free condition is set as the type 1, the obstacle in front is set as the type 2, the obstacle in left is set as the type 3, the obstacle in right is set as the type 4, the obstacle in front and right is set as the type 5, the obstacle in front and left front is set as the type 6, the obstacle in left, front and left is set as the type 7, the obstacle in right, front and right is set as the type 8, and the obstacle in left and right is set as the type 9, the obstacle avoidance control rule can be summarized into 9 types, and 189 control rules are set.
8. The automatic obstacle avoidance algorithm of the fire extinguishing robot according to claim 1, characterized in that: the structural design of the fuzzy control RBF neural network comprises the following steps: selecting three layers of RBF neural networks as neural networks of fuzzy control rules, taking input nodes of the RBF neural networks as distance parameters, taking output nodes of the RBF neural networks as speed parameters, and determining fuzzy magnitude values of input and output parameters according to membership functions of fuzzy subsets.
9. The automatic obstacle avoidance algorithm of the fire extinguishing robot according to claim 1, characterized in that: and when the defuzzification processing is carried out, defuzzification processing is carried out on the output of the fuzzy neural network, and the defuzzification algorithm adopts an area center method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110396197.5A CN113204235B (en) | 2021-04-13 | 2021-04-13 | Automatic obstacle avoidance algorithm of fire extinguishing robot |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110396197.5A CN113204235B (en) | 2021-04-13 | 2021-04-13 | Automatic obstacle avoidance algorithm of fire extinguishing robot |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113204235A true CN113204235A (en) | 2021-08-03 |
CN113204235B CN113204235B (en) | 2022-12-09 |
Family
ID=77026765
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110396197.5A Active CN113204235B (en) | 2021-04-13 | 2021-04-13 | Automatic obstacle avoidance algorithm of fire extinguishing robot |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113204235B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115576328A (en) * | 2022-11-15 | 2023-01-06 | 之江实验室 | Robot navigation obstacle avoidance method and device based on fuzzy controller |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH04324505A (en) * | 1991-04-09 | 1992-11-13 | Internatl Business Mach Corp <Ibm> | Autonomous moving machine and apparatus and method for controlling moving machine |
CN109272037A (en) * | 2018-09-17 | 2019-01-25 | 江南大学 | A kind of self-organizing TS pattern paste network modeling method applied to infra red flame identification |
CN110196596A (en) * | 2019-06-04 | 2019-09-03 | 南阳理工学院 | A kind of fuzzy barrier-avoiding method of four wheel mobile robots based on collision risk analysis |
CN111443603A (en) * | 2020-03-31 | 2020-07-24 | 东华大学 | Robot sharing control method based on self-adaptive fuzzy neural network system |
CN112631134A (en) * | 2021-01-05 | 2021-04-09 | 华南理工大学 | Intelligent trolley obstacle avoidance method based on fuzzy neural network |
-
2021
- 2021-04-13 CN CN202110396197.5A patent/CN113204235B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH04324505A (en) * | 1991-04-09 | 1992-11-13 | Internatl Business Mach Corp <Ibm> | Autonomous moving machine and apparatus and method for controlling moving machine |
CN109272037A (en) * | 2018-09-17 | 2019-01-25 | 江南大学 | A kind of self-organizing TS pattern paste network modeling method applied to infra red flame identification |
CN110196596A (en) * | 2019-06-04 | 2019-09-03 | 南阳理工学院 | A kind of fuzzy barrier-avoiding method of four wheel mobile robots based on collision risk analysis |
CN111443603A (en) * | 2020-03-31 | 2020-07-24 | 东华大学 | Robot sharing control method based on self-adaptive fuzzy neural network system |
CN112631134A (en) * | 2021-01-05 | 2021-04-09 | 华南理工大学 | Intelligent trolley obstacle avoidance method based on fuzzy neural network |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115576328A (en) * | 2022-11-15 | 2023-01-06 | 之江实验室 | Robot navigation obstacle avoidance method and device based on fuzzy controller |
CN115576328B (en) * | 2022-11-15 | 2023-03-28 | 之江实验室 | Robot navigation obstacle avoidance method and device based on fuzzy controller |
Also Published As
Publication number | Publication date |
---|---|
CN113204235B (en) | 2022-12-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108255182A (en) | A kind of service robot pedestrian based on deeply study perceives barrier-avoiding method | |
Patel et al. | IDUTC: An intelligent decision-making system for urban traffic-control applications | |
Khanum et al. | Towards a semantically enriched computational intelligence (SECI) framework for smart farming | |
Sharma | Designing and modeling fuzzy control Systems | |
Venayagamoorthy et al. | Navigation of mobile sensors using PSO and embedded PSO in a fuzzy logic controller | |
CN113204235B (en) | Automatic obstacle avoidance algorithm of fire extinguishing robot | |
Liang et al. | Collaborative energy-efficient moving in Internet of Things: Genetic fuzzy tree versus neural networks | |
Mehrani et al. | Sampling rate prediction of biosensors in wireless body area networks using deep-learning methods | |
Cavaliere et al. | Collective scenario understanding in a multivehicle system by consensus decision making | |
CN112949472A (en) | Cooperative sensing method based on multi-sensor information fusion | |
CN107967513A (en) | Multirobot intensified learning collaboratively searching method and system | |
CN112631134A (en) | Intelligent trolley obstacle avoidance method based on fuzzy neural network | |
Sun et al. | A game-theoretic strategy-aware interaction algorithm with validation on real traffic data | |
CN110827134A (en) | Power grid enterprise financial health diagnosis method | |
Gharajeh | Implementation of an autonomous intelligent mobile robot for climate purposes | |
Hossain et al. | Optimal fuzzy model construction with statistical information using genetic algorithm | |
Kong et al. | Path Planning of a Multifunctional Elderly Intelligent Wheelchair Based on the Sensor and Fuzzy Bayesian Network Algorithm | |
Díaz-Muñoz et al. | Cooperative Chaotic Exploration with UAVs Combining Pheromone Dispersion and Hopfield Chaotic Neural Network | |
CN116311950B (en) | Path selection method and V2X test system based on virtual-real fusion technology | |
CN110866777A (en) | Tourism big data business platform system | |
Yuan et al. | Method of robot episode cognition based on hippocampus mechanism | |
Feng | Multi-agent Collaborative Fire Rescue Based on Deep Reinforcement Learning | |
Fernandez | Predicting future network topology in wireless sensor networks with mobile nodes | |
Le | Fire Detection Robot using Type-2 Fuzzy Logic Sensor Fusion | |
Lin et al. | Type 2 Fuzzy Neural Controller for Navigation Control of an Ackermann Steering Vehicle |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |